tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 44683] connection pool is not being kept alive by dbcp
Date Sat, 19 Apr 2008 13:21:26 GMT
https://issues.apache.org/bugzilla/show_bug.cgi?id=44683


Marco Boniardi <marco@boniardi.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |marco@boniardi.org
             Status|RESOLVED                    |REOPENED
         Resolution|WORKSFORME                  |




--- Comment #2 from Marco Boniardi <marco@boniardi.org>  2008-04-19 06:21:24 PST ---
We have the very similar problem in our production environment.
Apache Tomcat/6.0.16
Jdk 1.6.0_05-b13 Sun Microsystems Inc.
Linux   2.6.18-53.1.4.el5 amd64
mysql-connector-java-5.0.8-bin.jar

This is one of our 6 pool:
    <Resource
      name="jdbc/bestprices"
      type="javax.sql.DataSource"
      initialSize="5"
      maxActive="15"
      maxIdle="15"
      minIdle="5"
      maxWait="5000"
      testOnBorrow="false"
      testWhileIdle="true"
      timeBetweenEvictionRunsMillis="40000"
      minEvictableIdleTimeMillis="240000"
      validationQuery="SELECT 1"
      username="bestprices_user"
      password="Abdftrf44"
      driverClassName="com.mysql.jdbc.Driver"
url="jdbc:mysql://vgdb05:3306/bestprices?zeroDateTime-Behavior=convertToNull"
    />

problems:
1) totally random the connection pool collapses, no errors, just hunged.
You can not retrive it from context with a lookup, we can not have connections.
All requests hunged. We think it's closely related to db idle connection
management, with idle connection closed server side due to timeout while the
connection pool can't see it.
2) We see validation query running, but very often twice than in
timeBetweenEvictionRunsMillis. We wrote 40 seconds, it runs every 80 seconds.
For some reason, in some machine of the cluster (older tomcat restart?) we see
validation query running 4 times (160 seconds) the time we defined, and it's
stable, always 160 sec., even if I don't know if tomorrow it goes to 320sec. or
it stays to 160. In our idea there is something wrong with the Timer that
manage query validation for idle connection. Problably
timeBetweenEvictionRunsMillis grows over server timeout period causing
connections pool collape.

Till 3 weeks ago we had Tom 5.0.x and everything was perfect, lower performance
but perfect. In these days we have to restart 1-2 servers /day in a 10 serves
cluster due to hunged connection pool. The problem is not related to traffic
because it happends night time too.

I hope this can help to figure out the problem.

Regards

Marco Boniardi


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tomcat.apache.org
For additional commands, e-mail: dev-help@tomcat.apache.org


Mime
View raw message